1915] Casimir Funk 25 



hence the figures are not exact. There was, perhaps, more sugar 

 in tumor serum, but not enough cases were investigated to Warrant, 

 a definite conclusion. Rolly and Oppermann found, in the few 

 cases at their disposal, an increase of sugar in the plasma.^ The 

 two sera which gave contradictory results were, strangely 

 enough, taken from animals in which inoculation occurred much 

 earher than in other cases. Whether these results were due to the 

 resistance of the animal to tumor growth, or to the weakness of the 

 tumor, must be left for further study to determine. For these two 

 sera the tumors were very small and entirely encapsulated. 



In summarizing the results the author feels justified in conclud- 

 ing that in the case of Rous's sarcoma, chemical analysis gives 

 much more reliable results than other diagnostic methods. Sev- 

 eral objections can, however, be put forward, The most important 

 of these is that Rous's chicken sarcoma differs in many respects 

 from other tumors; it is regarded by Rous himself as being of 

 infective origin. In a few instances rat serum was used — of rats 

 inoculated with Jensen's sarcoma. In each case 5-6 rats were bled 

 and the bloods combined, so that the figures in the Tables represent 

 good averages. Here, too, the differences were very much of the 

 same order as for chicken sera. This matter awaits further study. 



A second objection, no less important, is the fact that the same 

 chemical differences may be found in other pathological conditions. 

 The animals regarded as normal were brought up in town, and 

 were kept for a long time in the laboratory. Everybody who has 

 worked with fowls knows that, under such conditions, fowls do 

 not develop normally. Incidentally, a few cases of avian tuber- 

 culosis were investigated among the non-tumor animals and nor- 

 mal figures were obtained. 



We see, from the results in the accompanying Tables, that the 

 size of the tumor does not seem to have an effect on the data, 

 though possibly the length of the "inoculation period" played a 

 part. This point will be studied in the near future. The chlorin 

 content was found to be practically the same in tumor and normal 

 sera, and therefore can be disregarded. 



Although one is practically able to diagnose Rous's chicken sar- 



1 Rolly and Oppermann : Biochem. Zeit., 1913, xlviii, p. 471. 



